See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328253312

Empirical Correlations for Geometry Build-Up of Fixed Wing Unmanned Air

Vehicles

Conference Paper · October 2018

CITATIONS READS

10 1,928

6 authors, including:

Falk Götten D. Felix Finger

German Federal Aviation Office Airbus Defence and Space
28 PUBLICATIONS 265 CITATIONS 40 PUBLICATIONS 670 CITATIONS

SEE PROFILE SEE PROFILE

Carsten Braun Marc Havermann

FH Aachen University of Applied Sciences FH Aachen University of Applied Sciences
87 PUBLICATIONS 928 CITATIONS 34 PUBLICATIONS 330 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Falk Götten on 12 October 2018.

The user has requested enhancement of the downloaded file.

 2018 Asia-Pacific International Symposium on Aerospace Technology (APISAT2018)
Chengdu, China

Empirical Correlations for Geometry Build-Up of Fixed

Wing Unmanned Air Vehicles

F. Götten1 • D. F. Finger1 • C. Braun2 • M. Havermann2 • C. Bil3 • F. Gómez3

Abstract. The results of a statistical investigation of 42 fixed-wing, small to medium sized

(20 kg - 1000 kg) reconnaissance unmanned air vehicles (UAVs) are presented. Regression analyses are
used to identify correlations of the most relevant geometry dimensions with the UAV’s maximum take-
off mass. The findings allow an empirical based geometry-build up for a complete unmanned aircraft by
referring to its take-off mass only. This provides a bridge between very early design stages (initial sizing)
and the later determination of shapes and dimensions. The correlations might be integrated into a UAV
sizing environment and allow designers to implement more sophisticated drag and weight estimation
methods in this process. Additional information on correlation factors for a rough drag estimation meth-
odology indicate how this technique can significantly enhance the accuracy of early design iterations.

Keywords: Unmanned Air Vehicle • Geometry • Correlations • Statistics • Drag

Abbreviations

AR = aspect ratio L,l = length

bref = reference wing span LG = landing gear
CD0 = zero-lift drag coefficient MTOM = maximum take-off mass
Cfe = equivalent skin friction Ref = reference
coefficient S = area
cref = reference chord length TB = tail boom
cross = cross sectional UAV = unmanned air vehicle
D = diameter V = tail volume coefficient
EO/IR = electro-optical and infrared VT = vertical tail
FR = fineness ratio W = width
fus = fuselage wet = wetted
HT = horizontal tail

Falk Götten ()

e-mail: [email protected]
1 FH Aachen University of Applied Sciences, Hohenstaufenallee 6, 52064 Aachen, Germany
RMIT University GPO Box 2476, Melbourne VIC 3001 Australia
2 FH Aachen University of Applied Sciences Hohenstaufenallee 6, 52064 Aachen, Germany
3 RMIT University GPO Box 2476, Melbourne VIC 3001 Australia

1
 Empirical Correlations for Geometry Build-up of Fixed-Wing Unmanned Air Vehicles
APISAT2018

1. Introduction

The design and development of Unmanned Air Vehicles (UAVs) has come a long way since their first
appearance as remote control aircraft around 1918 [1]. With the significant micronization of electronic
components and the advances in computer technology, the UAV market has seen a significant growth in
the past years. UAVs are nowadays used for both military and civil missions including but not limited to
science, reconnaissance, agriculture or mapping [11]. Traditionally, engineers used their knowledge of
conventional aircraft design processes and adapted these to unmanned aircraft. This transition is ques-
tionable to some point, as the requirements and mission scenarios of unmanned and manned aircraft
differ significantly. Today’s state-of-the art UAVs have matured to an independent aircraft category with
very specific design properties [10].
A conceptual aircraft design process requires the analysis of several configurations and is an iterative
process with multiple refinement stages [8]. In the first stage (initial sizing), wing loading and thrust- or
power-to-weight ratio are commonly determined by using the so called “matching diagram” [7]. In the
second stage, the maximum take-off mass (MTOM) as a summation of empty mass, payload mass and
fuel mass is estimated by empirical regression analyses of similar aircraft. The geometrical shape of the
aircraft is determined in later stages using simple parameters like wing area, wingspan, aspect ratio,
fuselage length and so forth. Those parameters are subject to change during the iteration process. With
empirical regressions found for aircraft of a similar class, such parameters are initially estimated based
on a correlation with the maximum take-off mass. This initial guess is crucial for the following stages as
it can affect the number of iterations. Due to long experience with manned aircraft, empirical correlations
are available for most of the common aircraft classes and are presented, for instance, in Roskam [14].

The available data for UAVs is restricted to basic correlations necessary for an initial sizing process.
Several authors (Verstraete et al. [16], Finger [2] and Gundlach [8]) aim to provide such information.
Verstraete et al. [16] performed regression analyses of UAVs ranging from 0.1 kg to 40,000 kg MTOM
and developed correlations for payload fraction, endurance, empty mass, wing loading and power load-
ing. Finger [2] investigated empty mass correlations for UAVs between 2 kg and 1000 kg. Gundlach [8]
gives several correlations for basic sizing with a limited number of UAVs. With the currently available
data for unmanned aircraft, an initial sizing process can be established equally well for unmanned aircraft
as for traditional manned aircraft.

However, no data is available which links the UAV’s maximum take-off mass to more sophisticated
geometry parameters like fuselage length or landing gear size, to name only a few. The presented paper
aims to close this gap for small to medium sized unmanned aircraft by providing detailed statistical data
of 42 fixed-wing reconnaissance UAVs between 20 kg and 1000 kg maximum take-off mass. The corre-
lations can be used to perform a complete correlation based geometry build-up that provides a starting
point for further design iterations. Such correlations are of fundamental importance as they close the gap
between initial sizing, where only the basic parameters of the aircraft are defined (wing-loading, power
to weight ratio, MTOM), and later design stages in which the complete outer shape is determined.

This paper is structured in the following way: Paragraph 2 describes the data acquisition methodology
while section 3 presents the evaluation results, divided into several subparagraphs for individual compo-
nents. Paragraph 4 gives a distinct conclusion.

2. Methodology

The basis for the analysis is a detailed review of 42 reconnaissance UAVs ranging from 20 kg up to
1000 kg maximum take-off mass. This specific range was chosen as it represents the “small to medium
sized” unmanned aircraft category in which the general layout, propulsion systems and mission scenarios
 2018 Asia-Pacific International Symposium on Aerospace Technology (APISAT2018)
Chengdu, China

are similar. Very small UAVs (<20 kg) are often equipped with electric propulsion systems and feature
more special configurations like flying wings or blended wing bodies [1]. On the contrary, very large
UAVs can often be treated with correlations for commercial aircraft (see Verstraete [16]) and are there-
fore excluded in the presented analysis.
Correlations with geometry parameters can only be derived if a sufficient number of UAVs feature
the same components and if the aircraft configuration as well as the use case are similar. Only UAVs
which were at some point or are currently produced in significant numbers and have seen actual mission
deployment are included in the database. No experimental or technology demonstration aircraft were
analyzed. Such UAVs may never see mission use due to technical difficulties or changing requirements.
Additionally, such aircraft could also be designed exclusively for testing purposes of specific components
and it often cannot be determined whether instead of it they were sized to a specific design mission.
Excluding these aircraft increases the data accuracy for correlations of actual mission proven aircraft but
also limits the total number of UAVs.

An overview about all UAV configurations included in the study is given in Fig. 2.1, while detailed
information can be found in appendix Table 6.1. It was found that about 88% of all analyzed UAVs are
of tail boom or standard configuration. This paper therefore focuses on these configurations and neglects
more exotic ones like flying wings or blended wing bodies. “Jane’s All the World’s Aircraft - Unmanned”
[15] is used as the main reference for our research concerning the available UAVs.

This reference however, only provides basic ge-

ometry or mass information like wingspan, total
length or maximum take-off mass. To gain detailed flying wing 6%
information on individual component sizes, high-
quality images or three-view drawings are used,
twin tailbooms 51%
mainly provided by the UAV manufactures them-
selves. With basic geometry information, the dimen-
sions of nearly all components can be extrapolated
standard
from such images. Care was taken to include angu- configuration
lar- and distortion corrections to improve the data. 31%

Tests with UAVs where images and detailed CAD

models were available indicate that the accuracy for
individual component dimensions is in the order
of 6%. single tailboom 8%
blended wing body 2%
A spreadsheet based geometry break-down is de- delta wing 2%
veloped which divides UAVs into standard compo-
Fig. 2.1 UAV configurations in the database
nents and allows for an automated calculation of ge-
ometrical properties like cross sectional area, wetted area, aspect ratio and so forth. For this purpose, the
geometry is simplified and represented by shapes for which analytical equations are available. Depending
on the component, multiple geometrical representations are offered and the most realistic one is chosen
by the user. This simplification leads to slight deviations, especially for wetted area calculations. Taking
the data acquisition accuracy into account, the total deviations were found to have a maximum order of
about 10%, which is adequate for the desired purpose. Each UAV requires a manual input of at least 150
parameters on average. The complete workflow process is visualized in Fig. 2.2 while Fig. 2.3 shows an
example geometry simplification for a landing gear fairing.

The results are consolidated in diagrams (see paragraph 3) including all applicable UAVs. Those di-
agrams show the respective geometrical parameter in dependency of the UAV’s maximum take-off mass.
Regression analysis by means of a least squares fit is used to derive empirical equations representing the
best average of the available data.

3
 Empirical Correlations for Geometry Build-up of Fixed-Wing Unmanned Air Vehicles
APISAT2018

Image and drawing workspace

Image or Angular and distor- Dimension measuring with image

three-view tion corrections manipulation program

Spreadsheet workspace

Input in spreadsheet Calculation of dimension Data transition into predefined

database relationships and areas evaluation diagrams

Empirical correlations based on re-

gression analysis

Fig. 2.2 Data acquisition workflow process

Fig. 2.3 Example of landing gear fairing geometry simplification, image from [4] (CC0 universal public domain)

3. Results

The following section describes the analysis results of the study in forms of diagrams and their corre-
sponding regression analyses. The presentation follows the typical order in which UAV components are
sized and allows for a structured evaluation. Due to the enormous amount of data, the correlations are
reduced to the most important ones for each individual component. Wetted or cross sectional areas are
only provided for components if an accurate calculation is not possible by the already presented data.

Data scatter is more or less significant depending on the individual correlation. This is to be expected,
as the design space for UAVs is much bigger than for manned aircraft, which is a result of higher risk
tolerance, reduced costs and easier certification [10]. With an increased number of development compa-
nies comes increased variation in component shape and design. Conventional manned aircraft companies
tend to reuse technology from previous designs to decrease development costs and simplify certifica-
tion [12]. Such reuse of technology naturally leads to better correlations as several aircraft of one com-
pany might feature similar components. Within the market study it was observed that this does not hold
true for UAV companies. This again increases the expected geometry variations compared to manned
aircraft.

Even though some of the derived equations might show a relatively low coefficient of correlation,
they are still of high relevance for the design process of UAVs. The correlations enable the possibility to
estimate the size of UAV components given only the maximum take-off mass as an input. The equations
are trimmed towards the best average of the available data. This average leads to the most promising
 2018 Asia-Pacific International Symposium on Aerospace Technology (APISAT2018)
Chengdu, China

starting point for the design iterations in further development stages. The correlations might be integrated
into an automated sizing process (for instance shown by Finger et al. [3] ) and be utilized to perform a
complete geometry build-up in very early design stages. Such a geometry build-up can be coupled to
drag estimation methods or structural calculations and greatly improve the accuracy of the design feed-
back system. This consequently enhances the overall accuracy of the design process as it provides a direct
link between initial sizing and geometry design.

If not explicitly stated, all equations require maximum take-off mass as an input in kilogram and give
the geometry value in the unit stated in the corresponding diagram.

3.1. Wing

The diagrams below (Fig. 3.1 and Fig. 3.2) show correlations of the UAV’s wing area and aspect ratio
versus maximum take-off mass. Both regression analyses indicate a linear relationship and are presented
in Eq. 3.1 and Eq. 3.2. This is especially evident for the wing area, as it yields a rather constant wing
loading for UAV’s between 20 kg and 1000 kg.
All chosen UAVs feature reconnaissance mission scenarios which leads to similar requirements; thus
designers might aim at similar wing loadings. Data scatter for aspect ratio is significant, however, the
trend indicates that heavier UAVs tend towards higher values. The study also showed that larger UAVs
are very much trimmed towards extreme endurance missions. These mission scenarios require aircraft
with large aspect ratios increasing the achievable lift-to-drag ratio. Smaller UAVs might also be designed
for very high endurance, but most of them have a more diversified use case [1]. As such, the UAV is
more adaptable with a reduced aspect ratio. Additionally, with a small wing area, high aspect ratio wings
lead to short chord lengths, which affects torsional stability and could intensify aero-elastic effects. An
attempt was made to introduce the UAV’s endurance as an additional physical parameter into the corre-
lation but actually increased data scatter for smaller UAVs due to their diversified use case.

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000
14.0 14.0 25.0 25.0

12.0 12.0
20.0 20.0
10.0 10.0
SWing, m²

8.0 8.0 15.0 15.0

AR

6.0 6.0 10.0 10.0

4.0 4.0
5.0 5.0
2.0 2.0

0.0 0.0 0.0 0.0

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000

MTOM, kg MTOM, kg

Fig. 3.1 Wing area against MTOM Fig. 3.2 Aspect ratio against MTOM

Swing = 1.0339·10-2 · mMTOM + 1.1585 (3.1)

ARwing = 8.1658·10-3 · mMTOM + 8.7720 (3.2)

5
 Empirical Correlations for Geometry Build-up of Fixed-Wing Unmanned Air Vehicles
APISAT2018

3.2. Stabilizer Surfaces

Within the very first stages of an aircraft design process, stabilizer surfaces are commonly estimated by
employing empirically found tail volume coefficients, presented in Eq. 3.3 and Eq. 3.4 [7]. These coef-
ficients both incorporate the reference (wing) area and the reference chord or span as well as tail lever
arms measured from the wing’s quarter chord to the tail’s quarter chord.
Typical coefficients can be identified for specific aircraft classes. The present study found the average
horizontal tail volume coefficient to be 0.65, which is a value also common for homebuilt and general
aviation aircraft. Marshall et al. [10] found a slightly lower coefficient of 0.5 for small fixed wing UAVs.
Trends of the present study indicate that the coefficient slightly increases with increasing aircraft mass.
This corresponds to larger horizontal tail surfaces and increased stability.
The average vertical tail volume coefficient was found to be 0.042 comparable to the findings of
Marshall et al. [10]. The vertical tail volume coefficient shows a tendency to decrease with increasing
aircraft size, which leads to decreased lateral stability.

𝑉𝑉𝐻𝐻𝐻𝐻 ∙ 𝑆𝑆𝑟𝑟𝑟𝑟𝑟𝑟 ∙ 𝐶𝐶𝑟𝑟𝑟𝑟𝑟𝑟

𝑆𝑆𝐻𝐻𝐻𝐻 = (3.3)
𝑙𝑙𝐻𝐻𝐻𝐻

𝑉𝑉𝑉𝑉𝑉𝑉 ∙ 𝑆𝑆𝑟𝑟𝑟𝑟𝑟𝑟 ∙ bref

SVT = (3.4)
lVT

3.3. Fuselage and Tail Booms

Correlations of fuselage length and fuselage fineness ratio are shown in Fig. 3.3 and Fig. 3.4 together
with the results of the regression analysis in Eq. 3.5 and 3.6. Fuselage length varies from 1 m for small
UAVs up to 5.5 m for bigger ones. The length follows a power law trend, where the highest gradients
are found between 20 kg and 150 kg MTOM.
These observations are similar for the fuselage fineness ratio, which provides a link between diameter
and length. However, the regression analyses reveals that a logarithmic trend provides a better curve fit.
The fineness ratio varies especially between tail boom and standard configuration aircraft. Fuselages of
tail boom aircraft are naturally shorter as the necessary tail lever arm is provided by the booms. These
UAVs also feature smaller fineness ratios. The maximum fineness ratio was found to be 8.3 for a standard
configuration UAV, while the majority of fineness ratios is in the order of 4 to 6.
A correlation of the fuselage wetted area is shown in Fig. 3.5 together with the regression analysis in
Eq. 3.7. Fuselage wetted area is given here separately as an accurate estimation is not possible by
knowledge of length and fineness ratio only. Instead, fuselage wetted area is calculated by subdividing
the fuselage into four individual segments for which linear relationships are assumed. This yields to a
very accurate calculation. Fuselage wetted area shows a very slight power law trend.

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000
8.0 8.0 12.0 12.0

7.0 7.0
10.0 10.0
6.0 6.0
Fuselage fineness ratio
Fuselage length, m

8.0 8.0
5.0 5.0

4.0 4.0 6.0 6.0

3.0 3.0
4.0 4.0
2.0 2.0
2.0 2.0
1.0 1.0

0.0 0.0 0.0 0.0

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000

MTOM, kg MTOM, kg

Fig. 3.3 Fuselage length against MTOM Fig. 3.4 Fuselage fineness ratio against MTOM
 2018 Asia-Pacific International Symposium on Aerospace Technology (APISAT2018)
Chengdu, China

0 100 200 300 400 500 600 700 800 900 1000
20.0 20.0
Fuselage wetted area m²

15.0 15.0

10.0 10.0

5.0 5.0

0.0 0.0
0 100 200 300 400 500 600 700 800 900 1000

MTOM, kg

Fig. 3.5 Fuselage wetted area against MTOM

Lfus = 0.2825 · mMTOM 0.4206 (3.5)

FRfus = 0.7342 · ln(mMTOM ) + 1.6589 (3.6)

Swet-fus = 0.1219 · mMTOM 0.6773 (3.7)

The present study indicates that tail boom length is directly correlated to the UAV’s fuselage length
via a constant relationship. The results of the analyses are shown in Fig. 3.6 through Fig. 3.8, revealing
that the average tail boom length is about 0.75 x fuselage length. For the sake of completeness, a regres-
sion analyses of tail boom length against maximum take-off mass is shown Eq. 3.8. As tail boom cross
sections are mostly sized towards the acting loads, their diameter might be of higher interest than the
fineness ratio and is directly correlated with the maximum take-off mass (see Eq. 3.9).

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000
2.00 2.00 5.0 5.0
Tail boom length/fuselage length

4.0 4.0
1.50 1.50
Tail boom length, m

3.0 3.0
1.00 1.00
2.0 2.0

0.50 0.50
1.0 1.0

0.00 0.00 0.0 0.0

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000

MTOM, kg MTOM, kg

Fig. 3.6 Relative tail boom length against MTOM Fig. 3.7 Tail boom length against MTOM
0 100 200 300 400 500 600 700 800 900 1000
0.20 0.20

0.15 0.15
Tail boom diameter, m

0.10 0.10

0.05 0.05

0.00 0.00
0 100 200 300 400 500 600 700 800 900 1000

MTOM, kg

Fig. 3.8 Tail boom diameter against MTOM

7
 Empirical Correlations for Geometry Build-up of Fixed-Wing Unmanned Air Vehicles
APISAT2018

LTB = 0.5935 · ln(mMTOM ) - 1.2687 (3.8)

DTB = 0.01604 · ln(mMTOM ) - 0.00512 (3.9)

3.4. Payload

The payload of small to medium sized reconnaissance UAVs is normally an electro-optical and infrared
(EO/IR) gyro-stabilized gimbal pod attached to the lower side of the fuselage. Besides its overall impact
on weight and communication requirements, Götten et al. [6] showed that payload drag can be significant
for UAVs of the presented class and is largely driven by the respective cross sectional area.
Payload shapes range from spheres or half-spheres to combinations of spheres and cylinders. A geo-
metrical representation based only on their diameter is therefore only partially sufficient. Taking the drag
impact into account the cross sectional area is correlated against the maximum take-off mass, too.
Fig. 3.9 and Fig. 3.10 show that EO/IR payload size grows with the UAV’s take-off mass up to about
450 kg. Cubic functions give the best data fit and are presented in Eq. 3.10 and Eq. 3.11. For higher take-
off masses both payload diameter and cross sectional area stay nearly constant. This behavior was closely
investigated and found to have a distinct origin. Verstraete [16] found a constant empty mass and payload
fraction for a wide variety of UAVs, indicating that the primary origin for the findings in this study is
not the UAV’s structure or mission capabilities, but rather the payload itself.

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000
0.50 0.50 0.10 0.10
Payload cross sectional area, m²

0.40 0.40 0.08 0.08

Payload diameter, m

0.30 0.30 0.06 0.06

0.20 0.20 0.04 0.04

0.10 0.10 0.02 0.02

0.00 0.00 0.00 0.00

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000

MTOM, kg MTOM, kg

Fig. 3.9 Payload diameter against MTOM Fig. 3.10 Payload cross sectional area against MTOM

DPL = 1.2816·10-9 · (mMTOM )3 - 2.5110·10-6 · (mMTOM )2 + 1.5465·10-3 · mMTOM + 7.1638·10-2 (3.10)

Scross-PL = 1.9604·10-10 · (mMTOM )3 - 4.0708·10-7 · (mMTOM )2 + 2.8406·10-4 · mMTOM - 4.1643·10-3 (3.11)

A study of 43 EO/IR gimbals revealed that the packaging density of such gimbals increases with
increasing gimbal mass, as shown in Fig. 3.11 and Fig. 3.12. This density increase is significant and can
be as high as 100% comparing payloads between 1 kg and 100 kg. Heavier payloads therefore require
proportionally less volume than lighter ones. With a constant payload fraction, heavier UAVs also carry
heavier payloads, but payload volume and cross sectional area may stagnate due to the increase in pay-
load density. This leads to the fact that the aerodynamic impact of EO/IR payloads decreases with in-
creasing aircraft size. A critical region is identified for UAVs with maximum take-off masses between
150 kg and 400 kg (see Fig. 3.12). In this range, the payload’s cross sectional area is maximized in rela-
tion to the overall UAV size (represented by the reference area). The payload’s aerodynamic impact is
especially high in this region.
 2018 Asia-Pacific International Symposium on Aerospace Technology (APISAT2018)
Chengdu, China

0.1 1.0 10.0 100.0 0 100 200 300 400 500 600 700 800 900 1000
1600 1600 0.020 0.020
1400 1400
Payload Density, kg/m³

1200 1200 0.015 0.015

1000 1000

ScrossPL/Sref
800 800 0.010 0.010
600 600
400 400 0.005 0.005
200 200
0 0 0.000 0.000
0.1 1.0 10.0 100.0 0 100 200 300 400 500 600 700 800 900 1000
Payload mass, kg MTOM, kg

Fig. 3.11 Payload density against payload mass Fig. 3.12 Payload relative cross sectional area
against MTOM

3.5. Landing Gear

The layout of the landing gear is strongly dependent on the individual aircraft configuration and might
be influenced by factors like propeller ground clearance or center of gravity shift. Therefore, no attempt
is made to find correlations for positioning of individual gears or their detailed layout.
However, what is especially interesting for the designer are parameters like tire size or the landing
gear’s total cross sectional area. With this information, basic estimations of the landing gear’s weight
and drag are possible and might be used in the design feedback system. A power law correlation is found
linking tire diameter to maximum take-off mass (Fig. 3.13 and Eq. 3.12), while a linear relationship
between tire diameter and tire width can be identified (Fig. 3.14 and Eq. 3.13). Total landing gear cross
sectional area again scales linearly with the tire diameter (Fig. 3.15 and Eq. 3.14).
The last two correlations prove the common methodology of determining total landing gear drag as a
function of only one tire’s cross sectional area as valid for the investigated UAV class. This methodology
dates back to NACA Report 485 [9] and was also found to be applicable on UAVs [6], though the coef-
ficients found in the prescribed NACA report should be modified.

0 100 200 300 400 500 600 700 800 900 1000 0.00 0.10 0.20 0.30 0.40
0.40 0.40 0.16 0.16
0.35 0.35 0.14 0.14
Tire diameter, m

0.30 0.30 0.12 0.12

Tire width, m

0.25 0.25 0.10 0.10

0.20 0.20 0.08 0.08
0.15 0.15 0.06 0.06
0.10 0.10 0.04 0.04
0.05 0.05 0.02 0.02
0.00 0.00 0.00 0.00
0 100 200 300 400 500 600 700 800 900 1000 0.00 0.10 0.20 0.30 0.40

MTOM, kg Tire diameter, m

Fig. 3.13 Tire diameter against MTOM Fig. 3.14 Tire width against tire diameter

9
 Empirical Correlations for Geometry Build-up of Fixed-Wing Unmanned Air Vehicles
APISAT2018

0.00 0.10 0.20 0.30 0.40

0.30 0.30
Landing gear cross section area, m²

0.25 0.25

0.20 0.20

0.15 0.15

0.10 0.10

0.05 0.05

0.00 0.00
0.00 0.10 0.20 0.30 0.40

Tire diameter, m

Fig. 3.15 Landing gear cross sectional area against tire

diameter

Dtire = 0.04653 · mMTOM 0.28344 (3.12)

Wtire = 0.30489 · D𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 +0.00368 (3.13)

Scross-LG = 0.66490 · DTire +0.05981 (3.14)

3.6. Wetted area

The wetted area of an aircraft can be used to provide and initial estimate of the total aircraft’s zero-lift
drag by applying the equivalent skin friction method presented in Raymer [13]. Equation 3.15 shows the
method and the necessary input factors. Zero-lift drag is estimated by multiplying an equivalent skin
friction coefficient (Cfe) with the aircraft’s ratio of wetted area to reference (wing) area. The equivalent
skin friction coefficient is a constant for a specific aircraft class and estimated by regression analysis.
Factors for UAVs are not given in Raymer [13].
Fig. 3.16 shows correlations of wetted area versus maximum take-off mass as found in the present
study. Wetted area shows a strongly linear dependency on take-off mass described in Eq. 3.16. The ratio
of wetted area to reference area (Fig. 3.17) is mostly constant with the average around 3.8.
Götten et al. [6] provide zero-lift drag values for four reconnaissance UAVs of the small to medium
size class which were also analyzed in the presented study. Their findings are summarized in Table 3.1
and used to provide an initial estimate of the factor Cfe for UAVs of the presented class. Cfe varies between
0.00724 and 0.01163 with an average value of 0.00916. The difference between the highest and lowest
values is influenced by differing landing gear configurations affecting zero-lift drag [6]. Slightly different
configurations might significant affect zero-lift drag as outlined in [5]. It is noteworthy that the values
for Cfe are significantly higher than for every aircraft category found in Raymer [13]. It is evident that
zero-lift drag of UAVs is higher compared to other aircraft categories.

0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000
60 60 8.0 8.0
7.0 7.0
50 50
6.0 6.0
40 40
5.0 5.0
Swet/Sref
Swet, m²

30 30 4.0 4.0
3.0 3.0
20 20
2.0 2.0
10 10
1.0 1.0
0 0 0.0 0.0
0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000

MTOM, kg MTOM, kg

Fig. 3.16 Wetted area against MTOM Fig. 3.17 Relative wetted area against MTOM
 2018 Asia-Pacific International Symposium on Aerospace Technology (APISAT2018)
Chengdu, China

Swet
CD0 = Cfe · (3.15)
Sref

Swet = 0.04125 · mMTOM + 4.46529 (3.16)

Table 3.1 UAV data as given in Götten et al. [6] with calculated Cfe values
UAV Ref. area, m² Swet, m² Swet/Sref CD0 Cfe

1 1.32 4.76 3.61 0.03152 0.00873

2 2.37 8.57 3.62 0.04211 0.01163

3 4.74 17.26 3.64 0.03291 0.00904

4 8.70 36.16 4.16 0.03012 0.00724

4. Conclusion

The presented correlations allow an estimation of the size of the most important geometrical properties
of small to medium size reconnaissance UAVs by referring only to their maximum take-off mass. The
findings are valid for both tail boom and standard configuration aircraft with masses between 20 kg and
1000 kg. Regression analyses determined equations representing the correlations to a degree which is
adequate for early geometry estimations. These equations can be easily integrated into an aircraft design
and sizing environment. They provide a bridge between the very first design stage, in which power load-
ing, wing loading and take-off mass are sized and the following stages during which the shapes and
dimensions of components are designed.
The correlations might be used to perform both empirical drag and weight estimations already within
the initial sizing process and provide a valuable increase in accuracy in the flight performance estima-
tions. This enhances the accuracy of the design feedback system and thus the overall precision of the
sizing process.

11
 Empirical Correlations for Geometry Build-up of Fixed-Wing Unmanned Air Vehicles
APISAT2018

5. References

1. Austin R. Unmanned aircraft systems: UAVS design, development and deployment. 2nd ed.
Chichester, Hoboken: Wiley; 2010.
2. Finger DF. Comparative Performance and Benefit Assessment of VTOL- and CTOL-UAVs. In:
Deutscher Luft- und Raumfahrtkongress 2016.
3. Finger DF, Braun C, Bil C. An Initial Sizing Methodology for Hybrid-Electric Light Aircraft. In:
18th AIAA Aviation Technology Integration and Operations Conference; Atlanta. Reston ,VA:
American Institute of Aeronautics and Astronautics; 2018.
4. George Hodan. Ultralight Airplane. https://www.publicdomainpictures.net/en/view-im-
age.php?image=179889&picture=ultralight-airplane. Accessed 6 Jun 2018.
5. Götten F, Finger DF, Havermann M, Braun C, Gómez F, Bil C. On the Flight Performance Impact
of Landing Gear Drag Reduction Methods for Unmanned Air Vehicles. In: German Aerospace
Congress 2018.
6. Götten F, Havermann M, Braun C, Gómez F, Bil C. On the Applicability of Empirical Drag Esti-
mation Methods for Unmanned Air Vehicle Design. In: 18th AIAA Aviation Technology Integra-
tion and Operations Conference; Atlanta. Reston ,VA: American Institute of Aeronautics and As-
tronautics; 2018.
7. Gudmundsson S. General aviation aircraft design: applied methods and procedures. Oxford: But-
terworth-Heinemann; 2014.
8. Gundlach J. Designing Unmanned Aircraft Systems: A Comprehensive Approach. 2nd ed. Wash-
ington, DC: AIAA Education Series; 2014.
9. Herrenstein W, Biermann D. The Drag of Airplane Wheels, Wheel Fairings and Landing Gears I:
NACA Report 485; 1934.
10. Marshall DM, Barnhart RK, Shappee E, Most M. Introduction to Unmanned Aircraft Systems,
Second Edition. 2nd ed. Boca Raton: CRC Press; 2016.
11. Newcome LR. Unmanned Aviation: A Brief History of Unmanned Aerial Vehicles. Reston ,VA:
American Institute of Aeronautics and Astronautics; 2004.
12. Nicolai LM, Carichner G. Aircraft design. Reston, Va.: AIAA American Inst. of Aeronautics and
Astronautics; 2010.
13. Raymer DP. Aircraft design: A conceptual approach. 5th ed. Reston, Va.: AIAA Education Series;
2012.
14. Roskam J. Airplane Design. Ottawa, Kansas: Roskam Aviation and Engineering Corporation;
1987.
15. Streetly M. IHS Jane's All the world's aircraft - Unmanned: 2015-2016. Coulsdon: IHS; 2015.
16. Verstraete D, Palmer JL, Hornung M. Preliminary Sizing Correlations for Fixed-Wing Unmanned
Aerial Vehicle Characteristics. Journal of Aircraft. 2018;55:715–26. doi:10.2514/1.C034199.
 2018 Asia-Pacific International Symposium on Aerospace Technology (APISAT2018)
Chengdu, China

6. Appendix

Table 6.1 UAVs used in the presented study sorted by MTOM, (dash indicates lack of information)
No. Name Manufacturer MTOM, kg Payload mass, kg Endurance, h Length overall, m Span overall, m

1 Aerosonde Mk 4.7 Textron 25 - 14 1.70 3.60

2 Manta Raytheon 28 5 6 1.90 2.66

3 Luna EMT Penzberg 40 5 5 2.36 4.17

4 Sparrow EMIT 45 12 6 2.14 2.44

5 Atlantic SCR 45 7 6 2.80 3.80

6 Strix Aerodreams 48 18 15 2.90 3.60

7 S4 Ehecatel Hydra Technologies 60 - - 2.90 4.20

8 T-20 Arcturus 84 - - 2.90 5.33

9 Jump-20 Arcturus 95 - - 2.90 5.60

10 GRIF-1 558 ARP 100 30 8 3.50 4.80

11 Hermes 90 Elbit 110 10 15 4.20 5.50

12 Outlaw SeaHunter Griffon Aerospace 136 40 4 3.01 4.87

13 Pchela-1T Yakovlev 138 - - 2.78 3.25

14 Skylynx II BAE Systems 150 31 15 4.23 5.60

15 RQ-7B Shadow Textron 170 45 7 3.40 4.30

16 Shadow 200 AAI 170 27 6 3.40 4.30

17 Sentry HP Leonardo DRS 190 31 8 3.35 3.90

18 ZALA 421-20 Zala Aero 200 40 7 5.00 6.00

19 Pioneer RQ2-2A IAI 205 34 6 4.40 5.10

20 Tiger Shark NavMar 205 34 8 4.55 6.70

21 Aerostrar Aeronautics Systems 230 50 14 4.50 8.70

22 Pegaz MIT 230 40 12 5.40 6.34

23 Flamingo SATUMA 245 35 8 5.18 6.61

24 Shadow 600 AAI 265 41 14 4.80 6.83

25 Ranger RUAG 280 45 4 4.61 5.71

26 RQ-101 Night Intruder Korea Aerospace Industries 290 45 6 4.70 6.40

27 Xian ASN-209 Aisheng 320 50 14 4.28 7.50

28 F-720 UMS 360 70 12 4.80 7.20

29 Nishant DRDO 375 45 5 4.63 6.57

30 Searcher MK II IAI 450 120 15 5.85 8.55

31 Seeker 400 Denel 450 100 16 5.77 10.00

32 Falco Leonardo 490 70 9 5.25 7.20

33 Hermes 450 Elbit 550 150 20 6.10 10.50

34 Yabhon-R Adcom 550 100 27 5.00 6.50

35 Karayel Vestel 550 70 20 6.50 10.50

36 Atlante Airbus 570 100 15 5.47 8.00

37 Bayraktar Tactical Baykar 630 55 24 6.50 12.00

38 Falcao Avibras 630 150 16 5.90 10.80

39 GNAT 750 General Atomics 635 63 - 5.33 10.76

40 Rustom I DRDO 815 75 12 5.12 7.90

41 MQ-5B Hunter Northrop Grumman 885 113 12 7.01 10.44

42 Wing Loong CAIG 990 200 - 9.05 14.00

13

View publication stats